Shane A. Betman, Eldad A. Hod, Alexander Kratz, 57: Hydroxyurea Interference in Point-of-Care Creatinine and Glucose Measurements,American Journal of Clinical Pathology, Volume 143, Issue suppl_1, 1 May 2015, Page A030, https://doi.org/10.1093/ajcp/143.suppl1.031

Spurious iSTAT POC creatinine (and glucose) results with hydroxyurea

Many of us rely on iSTAT POC bloods some, or all of the time. We had a recent experience of an elderly patient who had a iSTAT POC creatinine of > 200 micromol/L and who we managed as AKI overnight. The next day his creatinine done in the main lab was 70 micromol/L.  Repeated tests  done on both the iStat and in the main lab using the same samples kept returning a similar  large disparity in creatinine levels. The problem in the end turned out to be the hyroxyurea that patient was on – which falsely elevates iSTAT POC creatinine levels. The manufacturer's advice when an patient is on hydroxyurea is ‘use another method’ to test the creatinine. And it looks like the hyroxyurea has the same effect on the iSTAT glucose reading and parcetamol (not at therapeutic levels but potentially in an overdose) might have the same effect of falsely elevating the iSTAT creatinine. Might be worth keeping this in mind.

Abstract

Background: Measurements of creatinine and glucose on the i-STAT point-of-care testing (POCT) device are known to be elevated in the presence of hydroxyurea. This interference can lead to differences between creatinine and glucose results reported from the i-STAT and samples analyzed with other methods. We sought to characterize the extent of this interference and to compare results with the epoc, a POCT device similar to the i-STAT.

Method

Patient serum samples with known creatinine levels were pooled to create three standards – normal range (NR), high (H), and very high (VH) creatinine. Serial dilutions of hydroxyurea were added to aliquots of each standard, resulting in final hydroxyurea concentrations between 0 and 2,000 μmol/L. Each aliquot was tested with the i-STAT, epoc, and Olympus platforms.

Results

Creatinine and glucose measurements on the iSTAT showed a dose-response relationship with the concentration of hydroxyurea in the sample. Disregarding data points outside the reportable range (output from i-STAT “>20.0” or “***”), the creatinine data fit linear regression models with slopes of 0.0138 (R² = 0.994), 0.0127 (R² = 0.995), and 0.0163 (R² = 0.978) for the NR, H, and VH standards, respectively. The glucose data fit linear regression models with slopes of 0.104 (R² = 0.999), 0.102 (R² = 0.999), 0.111 (R² = 0.998) for the NR, H, and VH standards, respectively. Creatinine and glucose showed no correlation with hydroxyurea levels when tested with the epoc or Olympus. All other analytes tested were unaffected by hydroxyurea levels.

Conclusions

Hydroxyurea causes linear dose-dependent elevations of creatinine and glucose results from the i-STAT POCT device. Based on our linear model and pharmacokinetic data, using the i-STAT following a typical dose of hydroxyurea could result in a creatinine level that is falsely elevated by 6.15 mg/dL on average and a glucose level that is falsely elevated by 46.09 mg/dL on average. Other platforms tested did not show interference by hydroxyurea. As the operators of POCT devices are unlikely to be familiar with the limitations of the testing methodology, it is important for laboratory professionals to keep them informed of appropriate practices.

© American Society for Clinical Pathology

NOTE: 6.15mg/dL (from the conclusion) = 543 umol/L and 46.09mg/dL glucose = 2.5mmol/L